RF & antenna engineer here. I’d like to know the input impedance at resonance. If you take two closely spaced, parallel dipoles and drive them out of phase, you get the same effect of nulling the far field. The issue is the impedance drops substantially.

Now take one dipole and insert a 1/2 wavelength delay (an additional resonator), and you end up with the same. So what they have done is synthesize the second dipole with the disk, so two dipoles along a common axis rather than parallel.

Dipole near fields drop off at 1/r^3, so yes, this is a method of near field coupling for power transfer or localized beating.