An observation - it lacks the third amplification stage the PR has right before the phase inverter which the cathodyne PI allows. ...
The Vox uses a LTP PI which has it's own gain so the overall gain of the Vox is still roughly equivalent to the gain of the PR with the gain stage plus the cathodyne.
Restating to make the similarity pointed out by SEL49 even more overt:
A typical 12AX7 gain stage has a gain of about 60.
A long-tail inverter delivers a gain
to one output about half what it would do in "a typical gain stage": so a gain of about 30 to one output.
A long-tail inverter obviously has 2 outputs, so we could say "the gain from one output to the other" is "30 + 30 = 60." 2 triodes deliver the gain 1 triode would normally provide.
A split-load/cathodyne inverter provides no gain of its own. It is composed of a single triode.
Almost always, a split-load/cathodyne inverter has a "pre-gain stage" right before it, and the feedback loop (when present) is returned to this stage (rather than to the inverter itself, as with the long-tail pair).
The "pre-gain stage" plus the split-load/cathodyne inverter now has "gain of 60" (pre-gain stage) plus "gain of zero" (inverter), matching the performance of the long-tail pair (at least in this manner).
The "pre-gain stage" plus the split-load/cathodyne inverter also is composed of 2 triodes, making it comparable to the long-tail pair in this manner.
Hopefully that helps explain why I said, "
The few differences are not super-relevant to me.". I probably should have said "troublesome" or "meaningful" instead of "relevant."
A difference that has very much more impact on the amp's sonics is the lack of a negative feedback loop around the power section in the Cambridge Reverb.