Yeah, I'd agree with that. I've found that very light, resonant, acoustically loud electric guitars feel "alive" in your hands but that doesn't always translate into good sustain.
Whereas guitars which feel quite "dead" acoustically can turn out to have really good sustain when amplified - often the case with hefty guitars like Les Pauls!
I can confirm that my Vox Custom 24, which is rather quiet (not very resonant) unplugged (I wouldn't say "dead" cause it still does react to playing, attack etc), has an amazing sustain even on pristine clean settings.
But the Vox Standard 25 I had a couple years ago had almost as much sustain, while being very resonant (and a strat-like construction where neither the bolt-on neck nor the vintage trem are known to help wrt/ sustain). The main difference vs a strat was the very thick and heavy maple body (which I guess was at least 4 or 5 parts - not supposed to help neither).
I sometimes think the resonance of the body dissipates the energy from the vibrating strings by sort of soaking it up, hence less sustain. That probably doesn't make much sense scientifically....
It does make sense and that's why we've seen all these very heavy neck-thru guitars in the late 70s / early 80s - which the Vox Custom 24 is a typical specimen of. But while there may be some truth in it, the whole phenomenon is obviously much more complex.
@Itamar: as everyone else already said, sustain (natural sustain I mean) depends mostly on mechanical factors, so if you feel like your guitar lost sustain then you know what to check for. You mentionned it had "gone through multiple changes with techs" - which of these changes could have had some impact on the mechanical side of things ? Anyway - this is IMHO a job for a good luthier (not "guitar tech").
Do you think a tubescreamer would help?
By definition, any clipping stage will level out the loudest part of the signal, which impacts the perception of the sustain - but this will not _add_ anything by itself. For "infinite sustain" you have to have enough gain and / or (preferably) volume to go into feedback, then find the correct distance / position to get this feedback on the right frequency.
