Abstract Moisture transport during food drying can be phenomenologically described by Fick’s second law and by the so-called anomalous diffusion model. However, in the literature, many studies have shown the extensive use of empirical/semiempirical models (EMs/SEMs) to adjust experimental data for the drying of thin-layer foods. This research aims to perform a critical analysis of the most commonly used EMs/SEMs and compare them with Fick’s second law and an anomalous diffusion model using two different sets of hot-air drying data. Two waste byproducts from the food industry, spent coffee grounds and passion fruit peels, were selected for analysis. The selected EMs/SEMs were found to be mathematically interrelated (i.e., some are a subset of others), and their appropriateness was incorrectly justified mainly by their statistical goodness-of-fit. As shown, it is highly recommended that researchers start analyzing drying data with phenomenological models. The extensive use of EMs and SEMs can be replaced by the anomalous diffusion model, which has a high capacity to adjust empirical data and a sound phenomenological description of the process.