The growth of Penicillium expansum and Aspergillus niger, isolated from yogurt production environment, was investigated on malt extract agar with pH = 4.2 and a w = 0.997, simulating yogurt, at isothermal conditions ranging from − 1.3 to 35 °C and from 5 to 42.3 °C, respectively. The growth rate ( μ) and (apparent) lag time ( λ) of the mycelium growth were modelled as a function of temperature using a Cardinal Model with Inflection (CMI). The results showed that the CMI can describe successfully the effect of temperature on fungal growth within the entire biokinetic range for both isolates. The estimated values of the CMI for μ were T min = − 5.74 °C, T max = 30.97 °C, T opt = 22.08 °C and μ opt = 0.221 mm/h for P. expansum and T min = 10.13 °C, T max = 43.13 °C, T opt = 31.44 °C, and μ opt = 0.840 mm/h for A. niger. The cardinal values for λ were very close to the respective values for μ indicating similar temperature dependence of the growth rate and the lag time of the mycelium growth. The developed models were further validated under fluctuating temperature conditions using various dynamic temperature scenarios. The time–temperature conditions studied included single temperature shifts before or after the end of the lag time and continuous periodic temperature fluctuations. The prediction of growth at changing temperature was based on the assumption that after a temperature shift the growth rate is adopted instantaneously to the new temperature, while the lag time was predicted using a cumulative lag approach. The results showed that when the temperature shifts occurred before the end of the lag, they did not cause any significant additional lag and the observed total lag was very close to the cumulative lag predicted by the model. In experiments with temperature shifts after the end of the lag time, accurate predictions were obtained when the temperature profile included temperatures which were inside the region of growth, showing that the assumption that μ is adopted instantaneously to the current temperature is concrete. In contrast, for scenarios with temperatures close or outside the growth region the models overestimated growth, indicating that fungi were stressed by this type of temperature shifts. The present study provides useful data for understanding the behavior of P. expansum and A. niger at dynamic temperature conditions while the developed models can be used as effective tools in assessing the risk of fungal spoilage and predicting shelf life of foods.