holy crap. i hope you're not planning on buying all this at once! this definitely looks a lot more well-rounded imo.
If you get the RCD and/or SCM i would highly recommend getting their respective breakout modules. it promotes the jumpers on the back of the module to front panel toggle switches.
i'm not 100% sure what you mean by an "on button" but there are a variety of power solutions available, and whether you need rack space for power is completely dependent on which one you use. i use the 4ms row power which is a 4hp module that uses a laptop power supply. it can also be daisy chained if you have more than one of them.
Anyone have problems with the Nano Rand being too tight in their case? I have Tiptop Z-Rails and the module's circuit board just barely fits in the space between the rails. This wouldn't typically be a problem but because the button is so close to the edge it binds up against the rail when i screw it down completely.
(it's also possible that my rail spacing is off -- my cases are homemade so i'm not 100% sure)
oh also it looks like you took my advice regarding having too many sequencers and removed a bunch of them -- which is good, but you removed all of the CV sequencers and you're left with all trigger sequencers. you're definitely going to want a CV sequencer of some sort.
nice, this is looking a lot more complete. you may not actually need the erbe-verb since clouds has a built-in reverb (and there's an alternate firmware that adds reverb as a processing mode); i would suggest passing on that at first to see if you really need it. i'm also not totally sure how much use you'll get from the lightplane, especially since you're just starting out, and it also takes up a lot of rack space.
i really like the disting. it does pretty much everything, and that makes it a great leading indicator of what you need to buy next (for example, there was a time where i was just using it for quantization, so i bought a quantizer. right now i'm basically using the delay mode exclusively, so i'm going to buy a dedicated delay module next).
it looks like you still just have the one filter -- you will most likely find this to be extremely limiting.
the only things you took out that i would think about keeping are the utility modules (links, ssf tool-box) and the turing machine, which is just a really cool module.
the tri-ger also looks really cool, that's definitely gonna be a must-buy for me.
if you start your journey with braids+clouds as your sound source(s) you will be far from basic :) i just got a clouds and already it's one of my favorite modules.
at first glance: it looks like you have a lot of redundancy where you don't need it. for example, i think you can get most of the functionality of the qu-bit nebulae with the clouds+radio music combo. you also probably don't need that many sequencers to start out.
do you really only have one filter? i would also suggest at least one or two more VCFs, an additional VCA, and an attenuator or attenuverter module (i know everything here sounds really boring, but it's the glue that holds your system together). VCA wise i would suggest getting something with a linear/exp switch, like the synthrotek VCA or the intellijel µVCA II.
you also seem to have a lack of modulation sources, like envelope generators and LFOs - i know that Peaks can do both of these but they're common enough that you probably want at least one dedicated module. consider a make noise maths, which can do A/D envelopes, LFOs, and a host of other neat tricks.
i like the look of that quantizer.
i believe the jones o'tool is not currently being manufactured, so you may have some difficulty acquiring one.
my overall advice would be to start small, figure out where the gaps are in what you want to do, and expand as necessary (e.g., one sound source, one filter, one sequencer, one modulation source, plus utility modules). if you go out and buy all of this at once you're going to find it really limiting because there are definitely going to be gaps (some of which are going to be completely dependent on the style you're trying to achieve).
