Sounds like you have a solid plan for an upgrade path and Tom might have some ideas for you. Here’s my two cents.
Do you have the schematic? If you’re on a budget you should target the parts in the feeds, rather than shunts, and prioritize the coax filters over the woofer (CS2.7 has electrical XO for the coax, right? CS2.4 coax is mechanical). That said, I was able to hear a difference among two voltages of otherwise identical capacitors in the woofer shunts, so *everything* should be assumed to make at least some sonic difference.
I recommend replacing as many resistors as possible on both boards as this makes a nice sonic difference for relatively little cost. Some of the CS2.7 values cannot be matched directly with MRAs, you will have to get creative with parallel and series replacements to get the correct resistance and power ratings. I noticed that PartsConnection has better pricing than Sonic Craft right now.
Clarity Cap CSA is a great sounding cap and a good value. Higher voltage has thicker film and this is desirable for the midrange where our ears are most sensitive. The downside is higher voltage caps are larger and more costly. CMR is said to be a step up from CSA but I haven’t heard these. You might also consider a bypass (1% or less total capacitance) on coax feed caps. I used Multicap RTX and heard an improvement in transients and, maybe, improved resolution. I was curious to also try CMR (Tom offered that as an option), Audyn True Copper Max, and/or Jupiter copper foil as bypasses. I dropped those plans after I put in the Cardas which just put the “cherry on top” of all the sonic improvements I heard along the way.
A couple of other things you might consider . . . 1) electrolytic capacitors will drift with age and this type is sonically inferior to film type, even if in a shunt position. The downside is replacing ELs with film is costly and requires far more space. You would need to make an entirely new board to replace that 400 uF EL with film caps. 2) Foil inductors are said to be an improvement over coils. I have ERSE FoilQ in the feed paths for both the coax and woofer boards. I did not make this particular sonic comparison, however, because I replaced the caps and inductors together.
Finally, Tom had cautioned me that such wholesale replacement of parts could upset the sonic balance in unpredictable ways and that one check was to compare the total resistance of the circuit before and after. I didn’t do this until after I replaced all the resistors but my final boards are within 2% of both the OEM boards with Mills MRAs and the values on my copy of the CS2.4 schematic (the CS2.4 woofer filter is almost a flow through, <1 ohm).