Whilst I certainly agree with Mike that the core is what makes for a good Sombrero, I can't help but to remain divided on this point. I recall Mike posting a Sombrero in the recent past and his insightful comments on processing, which have certainly stuck in my mind.
There are some interesting features local to the core which cannot be typically be resolved due to saturation therein, including a curious conical projection; an example of which I have attached. Perhaps someone can shed some light on the nature of this, as I cannot find any literature relating to it.
A side-effect of the local adaption filter, maybe?