I use six samplers playing the harmonic series n, 2n, 3n etc. to make an additive oscillator which is just like the modal synthesis description here: https://ccrma.stanford.edu/~bilbao/booktop/node14.html
If you ran adsrs into every osc, i reckon you could make a start at physically modeling sounds