Two different corncob samples from different continents and climates were studied by thermogravimetry at linear and nonlinear heating programs in inert gas flow. A distributed activation energy model (DAEM) with three and four pools of reactants (pseudocomponents) was used due to the complexity of the biomass samples of agricultural origin. The resulting models described well the experimental data. When the evaluation was based on a smaller number of experiments, similar model parameters were obtained which were suitable for predicting experiments at higher heating rates. This test indicates that the available experimental information was sufficient for the determination of the model parameters. The checks on the prediction capabilities were considered to be an essential part of the model verification. In another test, the experiments of the two samples were evaluated together, assuming more or less common kinetic parameters for both cobs. This test revealed that the reactivity differences between the two samples are due to the differences in their hemicelluloses and extractives. The kinetic parameter values from a similar earlier work on other biomasses (Varhegyi, G.; Bobaly, B.; Jakab, E.; Chen, H. Energy Fuels, 2011, 25, 24-32) could also been used, indicating the possibilities of a common kinetic model for the pyrolysis of a wide range of agricultural byproduct.