If the problem ONLY happens when another TX is transmitting, it must be an RX problem. Probably a sensitive receiver not filtering out the unwanted signals and then doing its best to resolve them. The simplest way is to substitute a known working receiver.
The crystals should be marked with their rated frequency, the TX crystal will show the frequency transmitted to air in MHz, the corresponding RX crystal will be offset by .455 MHz because thats how superhet receivers work. If the numbers match, and the pins are the right spacing, it does not matter whose radio they were made for. The value pf .455MHz is a common value for a single conversion RX, dual conversion is different and some manufacturers may have their own ideas.
Malcolm,
Agreed that 455KHz is standard offset for Single Conversion. It fits in with all the nice, off the shelf 455KHz filters. I'm not personally sure though, if (amongst RC manufacturers), there is a standard for whether the Rx oscillator runs at Tx Freq + 455Khz, or Tx Freq - 455Khz. Either permutation is equally valid for a superhet receiver, but would require slightly different circuit optimisation (as well as a difference in the Rx crystal frequency of 910KHz). If the Rx crystal is the correct value of offset (ie 455Khz) - but the wrong side of the Tx frequency, it may work, but if it does work, it may be problematic.
With regards to crystals, there are also other design parameters, which affect a crystal's resonant frequency. The resonant frequency for a given crystal is specified at a given load capacitance. If the load capacitance is different, the crystal will be pulled slightly off frequency, so (again, unless RC manufacturers have standardised these things), putting a crystal designed for one Tx or Rx, into a different Tx or Rx, will not necessarily result in exactly the same frequency.
My point with regards to FM vs AM is that crystals designed for FM Transmitters need to be lower 'Q' than crystals for AM transmitters usually are. The nature of Frequency Modulation requires that the Tx frequency is "pulled" by the signal. This is really only significant at the Tx end though, so not really relevant here.
I'm interested to know of any standards for these things - which may be used amongst RC manufacturers.
Ian