In the sense of routers (transmitters and receivers) it is preferable to consider the wave-model for the electromagnetic field.
The router has an antenna that, due to how it is shaped and how the current running through it is modulated, creates an electromagnetic field that propagates depending on direction and construction of antenna. For routers the ideal case would be to have an isotropic antenna (one that radiates equally in all directions) but this antenna is difficult to construct (a perfect isotropic antenna would be a metal sphere, but it has to be driven by current so this would destroy the perfect-sphere aspect of it) so wifi-routers are not radiating isotropically, but they're designed to be close to it.
When this electromagnetic wave produced by the router hits a wifi-receivers antenna, a current is induced in it. The phase and amplitude (and possibly frequency but this is rarer) of the electromagnetic field depends on the current running through the transmitting antenna, and this also creates a corresponding current in the receiver. The same current that is running in the transmitter is induced in the receiver, with more noise and lower magnitude however.
Your understanding of it as guided photons is inaccurate, they are not guided at all by each other but they are emitted in a directional pattern. These are waves that are travelling, and as any other wave, they surrender to the theory of superposition so they maybe interfere with each other in different spatial locations, but they are not "drawn" to each other in any way.