If the new 73-75 are exact replicas of the old systems there could be another issue - lack of buffering of sequencer outputs, causing a voltage drop when patching one output to multiple inputs. What this means is, you tune a sequencer row when patched to one oscillator (by ear or with a tuner), but then you decide to add a second oscillator, filter and so on, the first oscillator will no longer be in tune.
The CV Processor (1/2 of dual processor) can act as a buffer but this ties up a significant resource.
Alternately, a buffer can be patch-programmed using one section of the triple comparator, patching the comparator output to the “-” input and the signal to be buffered to the “+”, and with the offset knob at 0V. Instead of the comparator outputting a binary (0/5V) signal , it will act as a unity gain, with a buffered output at all voltages between 0 and 5V, which is generally the range of the sequencer output.
The reason this works is, the comparator without feedback is essentially an infinite gain applied to the difference of its two inputs, that clips between 0/5V. If you have negative feedback with a gain K, the overall gain is K/(1+K). Take K->infinity and result is a gain of 1 as long as the 0-5V limitation is obeyed. The comparator output is buffered and can be patched to many inputs without voltage drops.
In practice the gain is not really 1, or even linear, so you will have to tune the first oscillator directly from the comparator output instead of the sequencer row.
Again, I hope in the reissue they buffered the output (like modern Serge), but you can get around the problem with the comparator patch if this is not the case. There’s always this delicate choice, to offer more convenience, or to stick to the original circuits exactly – people complain in either case!
I think one can go far with this approach. A limitation is that one can’t transpose reliably by adding voltages. However there is more freedom in a sense. Anything now can be used as an oscillator with the same tuning procedure, for instance negative/positive slew, or filter patched in feedback, or some more complex feedback system. You don’t have to be constrained to the oscillators that track identically 1V/Oct.