The aim of this study was to model the effects of temperature (10–40 °C) and aw (0.80–0.98), in two media (Czapek yeast agar: CYA; corn extract medium: CEM) on the growth rates and growth boundaries (growth–no growth interface) of three strains of A. flavus isolated from corn in Argentina. Both kinetic and probability models were applied to colony growth data. The growth rates obtained in CYA were significantly (p < 0.05) greater than those obtained in CEM medium. No significant differences (p < 0.05) were observed among the three isolates. The growth rate data showed a good fit to the Rosso cardinal models combined with the gamma-concept with R2 = 0.98–0.99 and RMSE = 0.60–0.78, depending on media and isolates. The probability model allowed prediction of safe storage (p of growth < 0.01) for one month for moist maize (e.g. 0.90 aw) provided temperature is under 15 °C, or for dry maize (e.g. 0.80 aw) provided temperature is under 27 °C. Storage at < 0.77 aw would be safe regardless of the storage temperature. Probability models allow evaluation of the risk of fungal contamination in the process of storage, so the results obtained in this study may be useful for application in systems of food safety management.