The current materials are one of their greatest strengths, so I wouldn't be inclined to change that.
Cherry housings are super-powered compared with most other switches in terms of sound characteristics.
Tooling degrades too quickly - the first batches of a new line are typically good [not consistently], but then production becomes scratchy all-too-quickly. So they need to work on that. Refresh the tooling more, or use better tooling.
Springs are often noisy, and not that consistent. So that could be improved.
Tactile leafs are pingy until you break them in with thousands of key-presses. Can that not be improved?
Modern Hyperglide housings are designed improperly, they conflict with Cherry profile and certain stems. It's a problem that has most recently affected the updated MX Clears. The top housing in particular is the problem, I believe.
Wobble has always been a problem. Maybe copy OUTEMU's tight, no-slot tops from their custom-line.
So, this way, we might get regular Cherry switches, but with the old (Correct) housing design + quieter and more consistent springs + much tighter top housings + smooth stems.
Also, if the spring weight is more consistent, they could use 55-58 G on the MX Browns, instead of 60 G.
With regard to the Ergo Clears, they will need to come with more lube than is typical for Cherry in order to function well. At least, that's the case if the spring weight upon actuation is less than 60 G. With 67 G springs, you might get away with regular Cherry lube, might even be the case with poppy 62 G springs. But if using lighter 14mm 65 G springs, for example, which work well with Ergo Clears, you'd want some lube in there.
Not sure there is a point to factory Ergo Clears anymore, since everyone mods their switches and makes their own brand of Ergo Clears. Yeah, a 67 G factory Ergo Clear would probably work, but would anybody use it? They'd buy it to mod it, might as well use regular (pre-Hyperglide) Clears for that.