This test has been on the Bucket List for quite a while.
Bottom line, yes they do work, caveat: they don't work quite as well as an actual 2N404A but are easier to find, at least they were when I stocked up on them. Scope traces to follow and those will be good reference for any HV troubleshooting as well.
The difference in performance is negligible but recalibration is necessary, and that probably goes for any transistor swap out, but especially in V1 position - the HV generator. Just a small change in the HV setting will also change the -15V low voltage supply to the metering section, and that Voltage is key to the meter readings obtained. As a matter of fact this test led me to investigate some other stability problems in the ENi/LENi/Lionel CDV-700-6B.
Not surprisingly, battery Voltage when using dry cells is not stable over time. I jokingly call dry cells a 1/4 cycle AC generator. Compared to a real sine wave plot, the start at the + peak and go down to the zero crossover level, one time.
This changing Voltage was investigated, and sure enough, even a small change in applied DCV will alter the HV output and LV -15V output, causing changes in meter readings for a steady input pulse signal.
Geo
PS note the Blue and Yellow theme of the LENisky II PCB. A tribute to Ukraine.