We routinely deal with running quick checks of proof pressure analyses on manifold with SAE J1926 port or hydraulic valve ports.  We typically will apply pressure up to and including the sealing surface of the J1926 (as indicated by the red arrows in the bottom picture).  We do realize there is an imbalanced resulting force, but most times, these manifold are fairly "balanced" with ports on opposite sides and this is basically somewhat of a "sanity" check to make sure we are missing something drastic.  We do check this, and would adjust things accordingly.  My question here is the way our J1926 ports are modeled, specifically the threaded feature.  We have selected the "remove threads" options as shown here:

I am wondering if there would be any thoughts to the other options and if they would have better results?  Obviously, the option we selected results in a stepped model (red oval in picture below) which then gets pressure applied to the surfaces.  This is sort of "worse case" as the step is formed from the major diameter of the threads and the tap drill size.  So, basically I'm curious what others are doing for designs like this.  Thanks in advance...

This is a top view of one of the J1926 ports with a cross drill (green) and an additional axial drill (blue):