Well, I'm thinking what you don't want me to think: GET A BIGGER CASE. However, that IS the root of the problem here. That, and not thinking about the rig as a single instrument, which (invariably) comes up when these "fit a thing in X hp hole" threads appear.
First, consider a Mantis or something along those lines. Then, as you start to populate it, STOP thinking of the modules as individual bits of circuits. Instead, view what you're doing as setting up several blocks of specific activity...sound generators here, modulation there, filtering over here, etc. And make sure that you're putting those subsets together with modules that can synergize; you want modulation sources that can crossmod other mod sources, filters that can work in tandem, and the like. Then make sure that the ENTIRE build has that same synergy between the subset blocks. That's the real trick...and also, that's why most people can't just jump on MG and suddenly know what to do, as it takes time and effort to get to a point where that synergy factor gets optimized.
And yeah, it really does feel like the initial build above was the result of tossing things into a box. Case in point: the very overworked Maths, which is being pressed into service as an envelope generator, LFO, and probably a lot more other things. First of all, when you get dependent on one specific module like that, you're going to use it in ways that the dependency forces you to. The result here is sort of like buying a classic Lamborghini Countach for the purpose of going grocery shopping. Maths is capable of some VERY complex things...but not so much in this situation, because it's being pressed into the service of basic scutwork even though there ARE modules that can do those things better and more intuitively.