As to the RNAV glideslope that leads to a decision altitude that path often provides a more optimum TCH than does a VGSI. Also, in many aircraft the VNAV glideslope can be coupled, thus assuring a stablized descent path to DA if not the threshold. And, in marginal visibility conditions the VGSI can be indistinct in the visual segment, especially in the daytime.
Do you talk about baro VNAV here? Or similar concepts as Boeings IAN? If so, you need to be extremely cautious as baro-VNAV is not known for its precision, even half a hP is around 15 ft different TCH at sea level, more at higher airports. Since the reported QNH is only in full hPs and only every half an hour you can have quite a few feet of difference between the correct glidepath and the displayed one.
Granted, it looks very precise, especially if you have the same indication as on any precision approach (ILS/GLS), which is the standard on 737s now. However you have to be aware about the pressure and temperature effects on FMC derived glidepath indication. Something that PAPIs do not suffer from.