My money is on the camshaft as the cork in the bottle currently.
We will find out soon, I just sent in an order for one.
I'm of the same mind and if I'm wrong, a cam change is quicker/easier/cheaper than having the heads ported.
The existing camshaft has almost 102° of overlap and not near enough duration split for these cylinder heads. It may be a situation where the overlap is just pushing power out the exhaust. There are situations where the cam can be much bigger than what's needed, or even bigger than what the heads can handle, and the engine simply doesn't know what to do with it. I have ground smaller camshafts for guys in the past who were at wits' end trying to figure out how to go faster and have picked them up a tenth.
As an engine comparison, a 510" Tunnel Port here with much less duration (270/280 @ .050") and 380cfm heads will peak at 7000-7200. Jay's heads have 30+ cfm of flow over those.
Took me some time with a lobe catalog to find some lobes that were not spring eaters but had the lobe lift large enough to get us where we wanted to be with a 1.75 rocker ratio, while watching coil bind clearance with Jay's valvetrain.
If the camshaft doesn't let it zing on up, then I would look to the possibility of needing a little more intake port volume for a large engine. I'm running into this with the TFS heads right now, as they work really well with a ~170-175cc port and will support big horsepower on a smaller engine, but as I'm looking at some 496-505 ci engine builds, I think the port size will be a limiting factor. Got a set being ported right now to open up the envelope for a 496ci build going on.
If nothing else, we will get data, and data is always good, no matter which direction the trend is headed.