If it turns out not to be the most likely filter/oil-weight issue, a bit of gasket sealant of certain types could do some odd things in the oil gallerie's depending on what it has gotten into. If you have zero oil-pressure on the guage and the rod's are not pounding, then you're getting oil to the crank. So, I'm guessing oil pressure is read after the crank. On the gasket sealant issue, which is a fantasy in my head, it's propertie's in an oil gallery changing with heat and flow could conceivably do some very odd things like you're seeing. But, you are getting oil pressure to the crank when it reads zero, or you would know immediately. When it reads 100psi you might be getting a complete blockage by my mythical gasket sealant right after the point where the oil pressure is sampled, which then "flaps away" with coolant temp increasing causing little restriction. In the same fantastic vein a sealant goop before the pressure read out, but after the crank, would give a zero read.
I'd be interested to hear what your oil filter looks like if you'd cut it open for us, assuming that's the problem. I do think the filter being a problem is much more likely than oil viscosity. I believe a modern mechanical pump will pump 90W gear grease at 40 degrees F; I mean I wouldn't rev the engine til it's eeked it's way into all the machine fits, but just making a point. Oil viscosity does not seem to acc't for all of the behaviors you're seeing and their temporal sequence; though I say this in the relative sense, as a combination of viscosity and filter problems might do so.
I probably should read this before I submit it, but, I like contributing ideas more than I dislike being embarassed. One of our Maste Tech member's (DieselGiant ?) makes the beautiful point of how, if we are astute, we learn much deeper from out mistakes in thinking and application.