This paper presents an enhanced dynamic model of a self-excited induction generator (SEIG) with iron losses. The iron losses are represented in the induction machine model with an equivalent resistance connected in parallel with the magnetizing reactance. Since these losses vary with the air-gap flux and the stator frequency, a series of no-load tests were performed using sinusoidal supply of various frequencies to determine this variation. The air-gap flux influence on the iron losses can be expressed by means of the corresponding iron loss current. Consequently, an enhanced induction generator model, including the equivalent iron loss resistance as a function of both stator frequency and iron loss current, was built in the MATLAB Simulink environment. In addition, this is, to the best knowledge of the authors, the first SEIG model which includes variable iron losses that is entirely built in MATLAB Simulink. The performance of the proposed model is analyzed through simulation under various operating conditions and compared with the conventional SEIG model in which the iron losses are neglected. Theoretical results obtained from the proposed model have been verified experimentally. It is shown that neglecting the iron losses could cause severe inaccuracies in the SEIG analysis, especially when a SEIG is driven by a variable speed prime mover such as wind turbine.