I'm going to be a dissenting voice here. I have easily a dozen low power models, most BT-50-based, with over 30 flights each on them. I have two examples of one particular favorite of mine (Nova Payloader), one with 101 flights and one with 75, and my current flyer of that same design will go on its 33rd sortie the next time I go out to fly. Neither the 101-flight model nor the 75-flight model is very pretty any more. Each has been flown on up to composite Ds (and the one with 75 flights had several on the now OOP Aerotech D10-7s which have a blowtorch for an ejection charge...so the tube is soft and the paint bubbled above the motor mount, which is why it was retired at 75 flights though I think it is actually still flyable).
First of all, shock cords will fail and the "upgrades" suggested here will only put that off for a time. Just expect that it's going to happen (well unless you have a removable shock cord AND you replace it before it fails every time). Kevlar looped around the upper centering ring will burn through in ~35 flights. Estes' current white rubber, put in with the regular tri-fold (well away from the motor mount) can last that long if you use wadding correctly (and there's enough room in the model - things as short as the Alpha are harder). Inexpensive elastic from Hobby Lobby or Walmart won't last any longer than the white rubber either.
My current method for doing shock cords is to put a ~12 inch length of Kevlar line in with the good old reliable Estes tri-fold mount, then tie the original kit's shock cord (if it's recent and therefore long enough) or a good length of elastic (2x the model length at least) to that. I also have begun putting short length of 1/16th inch diameter heat shrink tubing over the kevlar where it rubs on the edge of my hardened-with-thin-CA top of the body tube (otherwise the edge of the tube will eventually saw through it).
I've yet to put enough flights on a model done this way to know how long it will really last. The 32-flight Nova Payloader I alluded to earlier has the most flights on this complete combination. So far no replacements of either shock cord or parachute.
But expect to replace the 'chutes after awhile....whether plastic or nylon.
I have never had issues with blow-molded plastic cones, and balsa, while it won't crack the way plastic will, does dent....so that's a trade of one issue for another. I use both....but I've never substituted one for the other in a given model.
As for beefing up the bottom end with epoxy or thicker/alternate materials for centering rings....I don't bother. The only epoxy in any of my low power or mid-power models is there to hold a screw-on motor retainer to the motor tube (24 or 29mm). For 18mm Beacon Fabri-tac does fine. Otherwise, it's all water-based glues or Beacon Foam-Tac/Fabri-Tac for plastic-to-paper joints.
Being aware of the balsa in the kit for the fins and if what is supplied is really soft swapping that out or maybe papering (I've never tried that) should help. One of the artifacts of a failed shock cord is often a hard landing for the body and a broken fin. But even just using wood glues to attach them, the fins don't necessarily come off, but break elsewhere. As I say, I've never tried papering fins, but that is one lightweight strengthening method that at least makes sense to me.
The other failure mode I often have to deal with is repeated strong ejection charges driving the motor aft and causing the top of the motor hook to tear the motor tube. Putting thin CA in the slot to harden it when building seems to help, but does not eliminate this. I normally do add a motor block above the top of the hook if the kit doesn't have one but that doesn't help the hook being driven aft by repeated ejection charges. This is hard to repair (though it can be done with care and thin CA with a long, thin applicator). I plan to use an 18mm Estes screw-on motor retainer on the next Nova Payloader I build to avoid this issue. Time will tell if this has me trading one problem for another.
Blast-it-Tom's suggestion about stuffer tubes in larger models flying on 18mm motors is a good one. My wife and I have built several Big Berthas and occasionally we have had simple failures to eject that probably would have not happened had we had a stuffer tube in the models as early Big Berthas did.
In general "improving" by adding weight, whether it is epoxy or just replacing parts with thicker/heavier ones just makes the crashes, when they come —and they will —harder.
I will be curious to learn about this "fireproof paper" that ZEDL1 speaks of.