First up, the uMod II is discontinued, so unless you have one or can source one, it's probably best to remove it from the build.

Secondly, don't do the drums in here! This build is TOO SMALL to support lots of different functions; you're already missing a lot of what should be there just to support basic synth functionality.

Third: Richie uses a lot of bespoke devices and software. And without these, you won't replicate his workflow. Not that you should, anyway...if you check your driver's license, and it's not from Ontario and doesn't say Richard Hawtin, you probably shouldn't be trying to replicate his setup. Given that there are SO MANY possibilities for individual builds in Eurorack alone, it's far easier to find your OWN sound than lifting someone else's.

Now, as for the build itself...first up, jettison the small cab. If you insist on using the Circadian Rhythms, well, that takes up a HUGE plot of panel real estate, and it would work better in a 2 x 104...not this. Next: drums...just get a proper drum machine. Something like a Digitakt, etc would actually be MORE flexible in the end than trying to sacrifice part of the modular for a function that's better left to a stand-alone device. And in fact, you SHOULD be building in a much larger cab to start with...then you should pare down from there to arrive at the right case form factor.

Next, utility modules aren't a luxury...they're a NECESSITY. If there's none of them (like right now, for example), then trying to get the maximum horsepower out of the expensive crap will be an exercise in frustration. If you lack VCAs, attenuverters, logic (ESPECIALLY if you're doing major sequencing!), and so on, your rig will suck like an Electrolux! And you WILL come to loathe it. FYI, you'll notice I'm not including the Invy here; $100 for a dual attenuverter/offset is a tad much, and you should be able to get into something with identical functionality for less.

As for things that're right here...the FH-1 is a very good pick, and frankly, you might find that sequencing in software and then piping that to the modular via the FH-1 would be a better choice. It definitely cuts back your space usage. And the White Gallop would make for a great internal clock gen/mod, but this would be improved by a couple of orders of magnitude by adding some Boolean logic so that you can massively screw with the sequential timing. As for the rest, though...this is more of a "do over". Try some of the suggestions above (especially expanding the case size), and see what you think.