It could be a good solution if reliable. But rebuilding only the translucent latch would help repairing the existant switches without having to change them.
there is a reason why there's such small actual variation in types available
making the latch pin would be viable though - but I'm rather sure that 3d-printing won't cut the cake. Wasn't here someone around here that experimented with this?