Fortunately, the engineers in the1980s were well aware of timing issues with optical transmission and designed simple circuits to render the issue (mostly) moot.
Those circuits appear to be intended to correct DC offsets in the optical receiver, and are to do with maintaining the correct duty cycle of the biphase coding, probably to ease data recovery when using an edge detector/MMV/DFF decoder. I don't think they will address the jitter, and, in fact, may introduce further timing jitter themselves.
Timing jitter is caused by a slow signal rise time and noise, either on the signal, or at the input thresholding device#. To minimise jitter, we need a fast edge signal, a fast receiver component, and low noise on signal and receiver.
Once edge jitter is introduced, it can only be removed by clock regeneration and signal reclocking. By measuring the jitter performance of modern DACs, you can see which DACs have clock regeneration and reclocking, and which don't.
# in fact, when developing a universal counter timer in the mid 1980s, we used an ultra-clean triangle wave generator to measure the input noise, using the variation (jitter) in the measured period, caused by the input noise changing the threshold voltage. Knowing the input slew rate and the rms jitter, you can calculate the rms noise of the input stage.
