There are probably multiple sources for interference, but they are getting into the MRF unit by only a few pathways: 1) the IR emitter cables, 2) the antenna picking up adjacent frequencies or harmonics from strong signals, or 3) the MRF unit itself picking up RF interference.

I would hope the design of the MRF would include sufficient EMI/RFI shielding to make #3 a non-factor.

In my case, it was all about the IR emitter cables. Ferrites worked just fine. But replacing the IR emitter cabling with twisted pair wire (as in cat5 cable) would also tend to reduce interference. There is also shielding in some cat5 cabling that could be grounded to further improve the noise immunity of the cable itself.

I think if you can unplug the IR emitter cables and the lights go off, then you're picking up the majority of RFI from the IR cables. But if the lights stay on, you are getting RFI induced into either the MRF unit itself, or the antenna is picking it up.

By touching the antenna, or twisting the wire around it, you are reducing the gain of (or attenuating) the antenna at the desired frequency. This also reduces the level of both the interference and the "good" signals from the remote. While this might not improve the overall S/N ratio, it does attenuate the absolute level of the "bad" interference.

People tend to think there is ONE problem, so there is ONE best fix. But sometimes there are multiple problems at the same time and the additive effects of them combined put you into this pickle. You might be able to reduce one of them sufficiently where it becomes tolerable.

Somewhere in the maze of home theater equipment, there may be some device that emits crazy amounts of RFI. Try powering them off one by one to see if the lights go out on the MRF - maybe there is one bad boy that is spewing out RF like crazy.