Radio-frequency (RF) circuits are crucial to qubit manipulation, for which transistor self-heating effects may influence performance and possibly change the quantum state. This paper presents an analytical RF model of FDSOI MOSFETs considering dynamic self-heating effects down to 3.3 K for the first time. Parameter extraction involves analytical calculation and optimization using the iteratively re-weighted least squares (IRLS) and Monte Carlo methods. The temperature rise is estimated by capturing the correlation between thermal resistance and device temperature. This work provides a method for modeling FDSOI RF performance and for analyzing dynamic self-heating effects at cryogenic temperatures.