In this paper, we present the results of simulating heat propagation processes in the thermoelectric detection pixel after absorbing single photons of UV radiation with energies ranging from 7.1 to 124 eV. The detection pixels, with an operating temperature of 1 K, consist of an absorber (W), a thermoelectric sensor (La0.99Ce0.01B6), a heat sink (W), and a substrate (Al2O3). The heat transfer processes in the detection pixels with different layer thicknesses and surface areas were modeled and simulated using the three-dimensional matrix method for differential equations based on the equation of heat propagation from a limited volume. The temporal dependencies of the temperature at various points in the detection pixels and the average temperature of the layers’ surfaces were calculated. The main signal parameters, such as the maximum value, time to the peak value, decay time, full width at half maximum, and signal power, were determined. In addition, the phonon noise, Johnson noise, and the total noise equivalent power of the detection pixels were calculated. The data obtained show how the signal-to-noise ratio (SNR) depends on the energy of the absorbed photon, the geometry of the detection pixel, and the bandwidth of the signal measurement. Moreover, it was observed that the SNR exceeds unity multiple times for the absorption of photons of all considered energies in the detection pixel with a surface area of 1μm2.