Wondering what the diode in parallel with the solenoid is for? It will forward-bias when it gets any juice, which happens when the diode in parallel with the relay contacts is bypassed by the coil energising and that won't IMHO leave much oomph left to energise the solenoid.
From what little info I could find of the Narva 68021 internal circuit, it appears to have a series resistance with the relay coil.
So to engage, 4th gear switch must be closed, and pressing "S" juices up the coil, which closes 30 to 87 and fires up the solenoid. But current will flow through the diode in parallel with the solenoid now that it's forward-biased, rather than just the solenoid. (Path of least resistance, y'know?)
How does the relay release..... I don't see how, since pressing "S" again juices up an already-juiced-up [86] and the cathode of the contact bypass diode again, which doesn't drop the juice to the coil, and that needs to happen in order for the contacts to open. Or maybe I'm missing something, but I don't think so.
JFTR, this is the Greenbook wiring diagram of the YM 1985 manual trans OD electrics, which has a momentary SPST switch on the gearlever to engage and disengage the OD, instead of the earlier slide switch. As this setup is for the late-model instrument cluster, the terminal numbers (and fuse numbers) will be different of course to the earlier models.
Interesting to note how the factory relay [H] has some sort of transistorised circuit within it to make it function like a step relay.
As an alternative to the Narva 68021, I wonder if a repurposed 240 headlight step relay might be another choice? The high beams engage and disengage when an earth is momentarily applied via the momentary SPST switch on the indicator stalk, and that's essentially what you want to accomplish for the OD solenoid via the pushbutton switch on the gear lever.
