Your intuition is correct. In fact, I'm not even touching this one; you need to rip at least half (or more!) of the signal sources out, for starters. This isn't a big build, and there's a hodgepodge of ten oscillators of various sorts (that I see on first glance!). Then there's ten channels of CV/gate sequencing, and another FOURTEEN of just trig/gate. But at the same time, just two filters...? If the idea here is to create a 'phat' (which I pronounce 'pee-hat', myself; shows you what I think of the term) sound, there's other/better/cheaper ways to do that. Or if you're trying to do something West Coast and crossmodulate everything...again, other/better/cheaper ways exist.
Step back and take a few breaths. Delete this. Start over, and start smaller. You're trying to toss things that look/sound cool into a box with this and figure that that'll result in a usable instrument, and it just won't. Seriously, go back and study some others' racks (synthesists who know what they're doing, basically) and spend a bunch of time studying classic modular systems that are still classics to this day because they were done right decades ago. See how and why they work. Then start again, but either smaller or with a smaller palette of modules (Eurorack, or also perhaps MU), or both, and then grow toward something like this incrementally. Build it up in subsystems that make sense as groups of modules. Use that grouping to place things in a sensible flow pattern.
Admittedly, this seems like it's easy. And on MG, it is...which may be the sole flaw in Modulargrid. Otherwise, it's a brilliant resource, but you have to use the whole resource to get the real use out of it. It's not Legos with knobs and wires; there's a real point to how and why modular synths come out in the ways they do, both good and bad builds. And the latter can and should (because this is 'expensive shit', to quote Fela Kuti) be avoided.