IMO, trebles are mostly a function of how well the top and it's bracing work together, while strong bass is more a matter of getting the rest of the box right. This is vastly simplified, of course.
You actually have to start out with the nature of the strings. Nylon/gut strings have a lot more damping than steel. Nylon and gut simply dissipate the energy of vibration faster; tapping on a piece of nylon gives a 'thud' sound, while steel rings on for some time. Add to this the fact that nylon strings , with their lower density, have to be thicker than steel to carry a reasonable amount of tension. The thicker strings have to move more air in order to vibrate. This doesn't produce sound: the strings are too thin for that, but it's like trying to run in knee-deep water. According to one calculation I've seen the two sources of damping are about equal in nylon strings.
Lots of damping tends to 'eat' high frequencies. If you put a nylon string and a steel one on a solid base, pluck them in the same place, and record the output (say, of a saddle pickup) you can see this. The initial wave form for the two strings looks exactly the same: the have the same proportions of energy in all the overtones. After a second or so the nylon string will have almost no energy above about 3000 Hz, but the steel string will still be producing sound out to 8000 Hz or higher.
This defines the central problems of making steel string vs nylon string guitars. For steel strings you're trying to get enough bass to balance out all the high frequency in the sound. You do this primarily by making the box bigger: enlarging the area of the top, making the ribs deeper, and enlarging the sound hole. All of this enables the guitar to 'pump' more air through the hole and off the top in it's 'bass reflex range'. The larger top, and higher string tension, require some 'beefing up' of the top, and they usually use a bracing system that is stiffer for the weight as well. Some steel string guitars use 'scalloped' bracing: the braces are actually scooped out lower at the bridge location. This makes the top particularly flexible in the center, which enhances the bass range output further, at the expense of 'sustain' and 'headroom', and an increased susceptibility to 'wolf' notes in the low range. Some of this is balanced off by using a heavy ebony bridge, and dense top wood.
With nylon strings you have the opposite problem: getting the most out of the small amount of high frequency energy you can get from the strings. This calls, in general, for a smaller top, which allows you to reduce the mass without losing stiffness too fast. The use of a low density top also helps. Long-grain Young's modulus in softwoods tends to track density pretty well in a reasonably linear way over the normal range of softwood densities. The stiffness of a top varies as the Young's modulus and the cube of the thickness: making a given piece of wood twice as thick will make it eight times as stiff, and adding about 25% to the thickness gives twice the stiffness. Leaving a low density piece of wood a little thicker to bring the stiffness up tends to result in a lighter weight of top.
Plucked strings have a limited amount of horsepower. If you want to make a car with a small engine, and you need to have good top speed and acceleration, you have to keep it light. Top speed in a car is the equivalent of sound power in a guitar, and acceleration equates pretty nicely with high frequency response. As I mentioned, large steel string tops need to be 'beefed up' to retain sufficient stiffness: with Classicals we go the other way: making the top span smaller allows for considerable lightening of the structure, and increases the ratio of top area to mass, which is what produces power.
Assuming you're sticking with a more or less 'standard' design, the key becomes building light enough to get decent bass power, while keeping the proper balance between the top and the bracing to maximize the high frequency response. A very thin top can be 'beefed up' with heavy bracing to get the overall stiffness up, but it will not respond as well at high frequencies. The relatively large unsupported areas between braces will be too floppy to work well at high frequencies. You can, of course, adapt by dividing the braces up: using more of them and making them smaller. This tends toward a Smallman -type system, where the braces take all of the load and the top is simply a membrane between them to move air. Another example of a similar system is a braceless 'sandwich' top, which is, in effect, a distributed I-beam: one large brace. In most cases the 'sandwich' type of structure is used as a sort of lighter-weight synthetic wood that is lower in surface density than normal wood, and is braced normally, but braceless ones have been made.
You could, of course, simply make the top thick enough to have the requisite stiffness, but that would make it too heavy. The function of bracing is to add stiffness without adding to much mass. It does this, of course, by being deeper than the thickness of the top; that 'cube rule' again, but not as wide. The problem with very tall braces on the thin top is, as I have said, that when the brace moves it can't 'carry' much of the top along with it if the top is not stiff enough: think of waving a flag on a pole. By the same token, a thin top can't move the brace very well, so vibrations of the top are pretty much confined to the areas between braces if the difference is too great. you run the risk of having lots of small areas vibrating out of phase with each other, canceling out any sound they would be able to produce. Another way to look at it is to see that a thin area of the top will tend to want to vibrate at a low pitch, while a stiff brace wants to go at a higher one. This suggests that the 'best' top for sound production might be one that has the stiffness of the braces and top well balanced over wide areas, so that neither dominates the other, or gets left behind, in terms of sound production.
There are a number of ways to do this. One of the best is simply to stick with proven traditional designs. A lot of the balance is, in a sense, already built in, by the trial and error process that produced them. The issue here is that wood varies: if you happen to get a really stiff top and some floppy bracing the balance won't be what it ought to be. Experienced makers learn to feel and/or hear the properties of the wood they use, and fine-tune things to get them to work right. Holding and flexing in various ways can tell you a lot; for example, by getting the lengthwise and crosswise stiffness of the lower bout balanced. David Hurd claims good results by developing deflection maps, and using them to fine tune the brace stiffness. A weight on the bridge of the assembled guitar pushes in the top, and the deflection is measured at a grid of points. Hard and soft spots produce bumps and hollows in the deflection map, and shaving the braces to eliminate those produces a 'better' top, in his estimation. Many people learn to use 'tap tones', remembering the sound of tops that work, and 'tuning' new ones to work in the same way. Many people learn to use 'tap tones', remembering the sound of tops that work, and 'tuning' new ones to work in the same way. A more modern 'tech' version of this is 'Chladni pattern tuning', which renders the resonant modes visible, and sorts them out by pitch. This can be applied to 'free' plates before they are glued to the ribs (when they're easy to get at!), and also to the assembled instrument.
Some makers find that thinning the top below the bridge brings up the basses, while leaving it thicker in the center and thinning out the 'wings' helps the trebles more. Bridge mass and stiffness can have a big effect, with a heavy bridge tending to favor basses over trebles, and vice versa. The thickness of the bridge wings is also a factor, as is, to some extent, the length of the bridge.
It's necessary in all of this to keep in mind that we're dealing with a very highly evolved design. In anything this developed the differences between 'average' and 'very good' examples tends to be quite small in an objective sense, but they are very important. Fine guitars are balanced on the head of a pin, in the sense that you have to get everything just right to achieve that outcome. I have seen examples of guitars where the sound was improved remarkably by removing a couple of shavings an inch or so long from some of the fan braces. It's also distressingly easy to start out with really fine material and make a mediocre guitar because you got one or two details 'wrong'. Keep in mind that, because of the variability of wood, 'right' and 'wrong' are moving targets: what's 'right' for one top might be 'wrong' for another that looks very similar. That's why fine guitars can't be mass produced.
Sorry for the long post.......