We all have opinions, I respect other's opinions, and this is my experience directly related to my XJ (FWIW).
K&N flat filter on my 88 4.0L, installed at ~30,000 miles (1989). The filter was washed and reoiled with a K&N filter charger kit about once a year (~20K-30K miles). Travels included weeks in Baja, severe dusty conditions, and extended service in the desert & mountains (my fathers ranch is all dirt road access).
This is what I found when I tore the engine down at 200,000 miles (1997):
Cylinder taper = ~0.002", no sharp ring ridge.
Valve guide wear = none to worry about.
Crank journal wear = rod journal clearances at ~0.0025" on the thrust diameter, and ~0.0020" on the perpendicular radius (oval journals, but within spec). The main journals were the same but ~0.0030" and ~0.0025".
The block was bored 0.030, the crank cut 0.010/0.010, and the stock valves reinstalled in knurled factory guides.
The engine would pass CA emissions tests with, or without, a cayalytic converter (I ran an empty converter in Mexico due to the poor availablity of unleaded fuel). The K&N made no emissions impact if clean (borderline emissions failure w/o cat when dirty). I experienced no negative emissions impact from the K&N oil.
The rebuilt engine now has 40,000 more miles, with a new K&N (the old filter is in the wife's 89), and burns no oil.
This wear is very reasonable for an engine under clean air street use conditions and 200,000 miles. I consider the wear very light wear for the extended 4Lo OD WFO-slogs in Baja sandwashes, Nevada dry lakes, and high desert & mountain ranch roads that I drove. I would sometimes find the airbox bolt heads buried in silt, but little downstream of the filter (other than the oil blowby that every hi-mileage 4.0L suffers).
I agree, a new paper element flows nearly as well as a new K&N, in an XJ application. The performance advantage is slight (2% at best). The airflow at 15000 miles is quite different, where the K&N has an advantage (IMO).
Does the K&N offer poor filtering? It filters less effectively (compared to a new paper element) when clean, but quickly matches the paper filter performance as it loads up.
Is the duration of time where the clean K&N is less effective a risk?
I don't know, from instrumented testing, but my experience is a 500 mile road trip after a filter cleaning is enough to provide adequate filtration efficiency from a K&N in severe conditions. This is what I would typically drive after cleaning the filter in preparation for a Baja pre-run, an LA area to San Felipe road trip before spending a week slogging the roads and washes between Ensenada and San Felipe (three times year minimum, from 88 to 96). If the filter failed to protect the engine when clean, I would have expected much more wear than what I found.
I have experienced no abnormal engine wear or emissions impact that would make me fearful of running a K&N (why I continue to run them, as opposed to changing a paper filter every 10,000 miles).
Anyone have experience to share that is similar, or different, from this (post K&N engine teardown analysis)?