Here are my two cents concerning EMI//RFI based upon my experience just for your information.
I doubt there any routers that are RFI free out there, but I agree that the quality or lower impact the device has upon its environment may be directly proportional to cost i.e. commercial routers may be less noisy and expensive versus the ones we can buy at BestBuy made for a house which are cheap and radiate a lot and may be affected from your signals as well. The difference between these devices is the specification that they are tested against of which there are as many as their are blades of grass in the universe.
Devices are tested to particular standards for EMI/EMC conducted and radiated emissions of which are based upon their intended end usage. As an example, devices used in the marine environment have specific requirements for EMI transmission within the areas of frequencies that other devices operate such as marine radio bands. A typical specification is the ABYC P-24-07 which indicates that between 156MHz and 165MHz a device can not radiate more than 24 uV/m of energy, protecting the marine VHF band. There are other levels within that specification between 150kHz and 30MHz, also, although the radiated levels are not as low as for the VHF. The ABYC spec is primarily used with respect to marine radio interference and I doubt that routers are tested against that standard. So routers we purchase for cheap may not be tested with reference to ham radio frequency spectrum so unless you can see the results and know the standards said routers are tested to--most likely relates to other commercial devices on the premise of which ham radio gear most likely is not on the list, then you have to resort to trial and error to reduce their effects on your radios - a spectrum analyzer is handy for experimentation as long as you know what the noise level is of the environment and work from that reference. Not sure if those can be found at a hamfest or not
Just so you know, every electronic device that operates in automotive, marine, and medical environments are tested for EMI/RFI to any host of governing bodies with respect to specifications. Specifications range from those developed by automotive suppliers, SAE, IEC, IEEE, DOD, and others with most of these referencing back to some larger regulatory standard such as IEC. I work with many of these and part of the challenge is to find one that works with all of them within my industry of transportation.
Radiated emissions are influenced by many things and primarily with anything of course that "vibrates" such as power regulators taking Vbat down to 3.3 or 5VDC levels. A typical fix for a noisy regulator on the board level may involve moving components away from noise producing devices and/or adding very tiny chokes on conductors or adding guard traces to the board. Placement of the devices with respect to other pieces of hardware (and its wires) that may be affected by it can be an easy way to reduce a radiating devices effects (remember those notifications on the back of electronic devices?) plus one can add chokes to lines, reduce lengths, and all that. Even how the wires are bundled outside the device affect how much/efficient it can radiate.