I have some suggestions that may help.  I am a composite designer/builder and also design insulation and interiors for aircraft in addition to building an ES.

First off, if you repair the cowling in the area of the damaged core, you will have some shrinkage there and depending on how anal you are, may have to fill, prime and repaint the area externally.  So be prepared for some body work afterwards.

Next, DO NOT USE STANDARD EPOXY FOR THE REPAIR!  The heat distortion temperature is WAY TOO LOW for this.  The prepregs used were for 260 degrees if I remember correctly and the Jeffco epoxy is designed to have an HDT (heat distortion temperature) of 190 degrees.  Use a high temperature epoxy.  I like Applied Poleramic in Benicia Ca. (707) 747-6738.  Talk to Rich, the owner and head chemist for specifics but he only makes high performance toughened epoxy.  Toughening makes the resin a lot higher bond strength and tolerant of impacts.  And interestingly, he still doesn't charge exorbitant amounts of money for it.  Anyway, adding 6 layers of BID is probably adequate but you haven't described how large the area is.  I'm assuming it's small so I'd tend to agree but as with all laminates, taper the layup so that there is a gradual transition from the thickest to the thinnest layers and also alter the orientation of the fibers.  Note that by altering the orientation of the fibers, you disperse impact energy better so this is something to keep in mind for other more crucial laminates.  For this repair it's not really all that important since you are designing a repair for stiffness, not impact.

And for the repair, taper (with a grinder) the area that is disbonded and damaged.  All of the damaged area should be ground away and the tapering should begin in the good area. This should also improve the airflow locally when compared to hard corners. Do NOT do all of the layup at once.  Start with your second largest piece of glass and wet it out on a wet out board and DAMPEN the tops of the cells of the honeycomb.  Let this harden but NOT CURE.  Grind a little taper at the edges to facilitate smooth edges to layup over plus hit any nibs, then going from the smallest layer to the largest, layup the fabric over the previously layed up  fabric.  The last piece should cover everything and will give a smoother look plus superior structure since it avoids stress risers due to tapering. The reason you want to do it in  two stgaes is that if you do it in one shot, the resin drains into the cells and makes for a sloppy, heavy layup that tends to be weaker since the fabric may be left with insufficient resin.  Put a heat lamp on it after it has hardened to post cure it but don't over do it!  You could also go out and fly it and the temperature from the turbo will cook it.  The thing to remember about high temp resin is that while it will harden at room temp, it will not achieve its full structural integrity until it is post cured.  This means that you could do a perfect repair and refinish on the outside that looked great, but the next time it was heated, it would shrink some more and show the repair.  It would continue to do this until it peaked on the heat it sees.  Structurally, you are now done.

For those that wanted to, you could also grind out the core and replace it with core similar to the original.  Just router out the bad core, grind the bad area to prep for bonding, dampen the core edge and bond in place with Cab-O-Sil thickened epoxy.  Then grind smooth the contour of the core (once the core bonding epoxy is set), and laminate. You would only be needing a couple of layers of BID over it but again, use high temp resin and also you need an insulation blanket.

The company I work for (Orcon Corp.) can supply you with high temp insulation blankets.  We have some really great stuff that we use for business jets that demand high performance.  For this application, I believe that an aluminum foil covered blanket 1/4" thick should get the composite surface temperature down significantly enough to prevent the skins from being damaged.

If someone out there is willing to do some experiments, I can send a test blanket out and some temperature probes to see what the actual temps are.

Contact me at work at (510) 476-2150 or by email prewferably at work at dan.newland@orcon.com