The screen is so spacious, I suspect you are right about this direction.
That was the idea, yes.
I can see how this would make a polyphonic patch more organic and expressive, but it would also make it less predictable (unless, perhaps you had a really strong mental model of how voice allocation is working and can anticipate the variations? I don’t yet understand the voice allocation at this tactile level).
Yes, perhaps copy/paste addresses the need, but maybe store the idea of “number of voices” parameter away for later. Might eventually come in handy for more complex polyphonic patches.