A comprehensive mathematical model of wood pyrolysis, based on the unreacted-core-shrinking approximation, is used to assess the role played by the description of the kinetics for a one-step global reaction. The more accurate model, which was previously subjected to experimental validation, includes a first-order Arrhenius rate, whereas the simplified one uses the assumption of constant (assigned) temperature, T-p, at the pyrolysis front. Both models predict qualitatively similar particle dynamics. Extensive simulations, carried out by varying the parameters of the kinetic models, the external heating conditions, and the particle size, indicate that the unknown parameter T-p, though comprised in the range of experimental values, should not be chosen coincident with the pyrolysis temperatures predicted by the finite-rate model. However, a range of T-p values can be determined that produces chief process characteristics, such as maximum devolatilization rate and conversion time, very close to those of the finite-rate model. In this way, acceptable agreement is also obtained between predicted and measured integral and differential mass loss curves of thick wood exposed to radiative heating.