My understanding is the Narziss cold and secondary profiles are designed to keep yeast active longer and colder versus a fast drop. Nico did a nice writeup on this: https://edelstoffquest.wordpress.com/20 ... mentation/
As ectothermic organisms, the yeast will have a faster metabolism at 3C than 0C, and yes they will clean up and condition the beer faster. The risk to lagering at 3C is the increased potential for yeast lysis, as they will consume their reserves faster than at 0 or -1C once they flocculate. Some have even claimed that the increased metabolism can consume some of the flavor compounds and foam and mouthfeel proteins that you are trying to preserve. Like with most things, there is always a compromise.
That said, one of the temperature profiles I use is the most is 6->9->3->0C. Pitch at 6C, rise to 9C, spunden and condition at 3C, umdrucken (or not) and lager at 0C. I find it cleans up the beer faster than the Narziss secondary profile, but not by much, and then I can really put the yeast to sleep for long term lagering and slow down how quickly the yeast die off. Umdrucken or filtering is clearly the best option for long term stability.
My personal experience is that you can go too cold, too early, especially if you do it too quickly and shut down the yeast. I've had beers that tasted clean, then had slight diacetyl, then were clean again. I suspect you can actually produce a situation where the yeast are cleaning up the VDK's faster than the precursors are converting to diacetyl, as yeast conversion is enzyme regulated, whereas precursor conversion is driven by simple thermodynamics. I've wondered if this is part off the trick (accidental or not) to getting a consistent level a diaceytl in the Czech styles. The yeast strain is important too, as the Czech strains tend to throw 3-5x more diacetyl than most other lager strains.