This is typical, you need to be concerned with the mash temp and not the liquid. Adjust your digital temp up or down to get your analog thermometer in your mash to your desired temp.
The liquid portion of the mash is not just a fluid whose sole purpose is to heat the grain in the tun, it is also a significant portion of the 'mash'. You have to consider the 'mash' as a whole. That is why nobody with any sense uses a probe in the grain bed as the input to the temp controller for a RIMS tube or HERMS heat source. This is a direct fired RIMS MLT which is even more problematic from a temperature gradient and differential perspective.
That said, even for a steady state mash profile, there will always need to be some temp differential between post-RIMS/HERMS and the tun as long as there is heat loss in the tun and system, which there always will be. However, there is a limit to the amount of heat that can be infused into the fluid to counteract the heat losses. The magnitude of the differential is a personal choice based on your beliefs regarding mashing science, enzyme denaturing, etc., but there is a limit. A 7F diff is on the high end of that range, even for a HERMS or RIMS tube, unless your mash temp tolerance is > 5F.
The situation at hand is even more complex because it is a direct fired MLT RIMS instead of a more gentle HERMS or even PID controlled ULWD RIMS tube. With a direct fired MLT, there will be a layer of higher temps close to the heated kettle surface. There is also no easy way to get a temp reading close to the pre-RIMS side of the mash to know what the mash exit temp really is. Additionally, and this is the most important part, the tun temp is coming from a slow reacting complete POS analog dial thermo that only probes in a few inches into an isolated pocket of the grain bed very close to the tun wall which is the biggest heat sink in the whole system and gets more or less exposure to the recirc flow-through depending on where the single-downward-pointing-hose-barb-in-the-lid based sparge arm (jet?) is currently boring a vertical hole in the grain bed.
Insulating the tun would help significantly, but it is direct fired so it is problematic to find a flame proof way of insulating it.
Have you considered going to a HERMS?
Only this time try one that isn't based on your wife's $10 digital turkey thermometer that reads 40F low; is only noticed 40 minutes into the mash when it quits completely; and results in an IPA that looks like a pint of the water after the 'chefs' at Golden Corral finish cooking up the day's trough of pasta. And no, I didn't just make that up.