Hi,
Going through a .czi metadata to extract some information I ran into both the NA 0.95 of the lens and a value set as the effective Numerical aperture a smaller value at 0.8. I wonder if anyone here can help me find this difference and from there to know which one theoretically is the one to be considered in the computatation of the point spread functio.
Many thanks! 
Hello and welcome to the Forum!
I cannot give you a definitive answer but an opinion. This is because Zeiss ZEN is very much NOT open source software and you need to get a license just to get access to their file format details but maybe someone here already has this or knows the answer more definitively.
Having said that, from a purely scientific point of view (and it would be interesting to learn if the ‘official’ Zeiss .czi format definition differs from this), the ‘effective’ numerical aperture would be the angle of acceptance of illumination (refractive index of the medium taken into account as per the Abbe formula) that was permitted during the experiment regardless of the maximum NA specs of the lens.
The NA of the lens is the maximum NA rating it can practically handle using the immersion medium for which is was designed (air, water, glycerol, DIN oil, etc.). You can easily reduce this in practice for any experiment by (e.g.) stopping down the condenser iris and/or using a different mounting and immersion medium to the one for which the NA of the lens was measured at during the lens design and manufacture stage (e.g. using water or glycerol immersion instead of DIN oil with a lens designed for DIN oil immersion will reduce the effective NA of the experiment with that lens to below its rated standard).
I much prefer open source software, hardware and documentation for scientific transparency. The companies can still make money selling their expensive stuff - open source doesn’t mean they have have to give their kit (or software) away for free.