I'd just point out one fact that many people miss.
If you are recording stars as points of light, it's the absolute physical aperture of the lens that counts, not the relative aperture (fstop), when it comes to the exposure needed to record a star of a particular brightness.
This is counterintuitive to many photographers, but it's because stars are point sources, not extended objects. If you are photographing extended objects like nebulae, then the f-stop does count. The f-stop also influences the amount of "skyglow" that you capture, i.e. how long it takes before the sky starts to "wash out".
Obviously the physical aperture of a 16mm f2.8 lens is pretty small (5.7mm), so though it may record the sky glow (and aurora) pretty fast, a fairly long exposure may be required to record faint individual stars.
If you pixel peep, the edge quality of the Zenitar wide open isn't going to be diffraction limited I'm afraid, but I'm not sure any lens will be. Stars are a very severe test of optical quality