Hi, I was looking at this too for a time. For starters i found you need to make sure windows are added in the workflow before you add the window control components to the model. Secondly I think a lot will depend on the type of windows you are testing, the window controls, and whether the room has other energy loads e.g HVAC running.
You might find there is a difference in the room air temperature even if you see none in the MRT. For example, I found when testing a simple box room with 90% glazing on all four walls, made of the ‘Generic Single Pane’, and no other internal loads (i.e it is plugged into the plenum zone component), there was about 1C decrease in MRT with windows closed and a 6C increase in air temp.
I assume you’re using EnergyPlus and not the UTCI component? You might want to check this thread on how EnergyPlus calculates MRT to give you more info: Mean Radiant Temperature from Energyplus results?
The UTCI component could be quite useful for you in understanding how the MRT is affected by your windows because you can separate the effect of the solar radiation (shortwave MRT) and the effect of the surrounding surface temperatures (longwave MRT).