Aims The non-rectangular hyperbolic model (termed as model I) is the main submodel of the FvCB biochemical model, which is used to estimate the maximum electron transport rate (J max) of plant leaves. The submodel is widely applied to fit the light-response curves of electron transport rate (J-I curves), and obtain J max. However, it has not been strictly verified whether J max calculated by model I is consistent with the measured values. Methods Light-response curves of electron transport rate and of photosynthesis rate of soybean (Glycine max) (under shading and full sunlight) were simultaneously measured by LI-6400-40, then these data were simulated by model I and the mechanistic model of light-response of electron transport rate (termed as model II). Important findings The results showed that there was the significant differences between J max estimated by model I and the observation data irrespective of shading and full sunny leaves of soybean. However, there was no significant difference between J max calculated by model II and the measured value. Because J max was overestimated by model I, it must lead to overestimate the amount of photosynthetic electron flow to allocate to photorespiration pathway, and magnify the photoprotection of photorespiration on plants. On the contrary, the J max and saturation light intensity (PAR sat) obtained by the model II were in very close agreement with the observations. It can be concluded that the model II was superior to the model I in estimates of J max and PAR sat. Therefore, we recommend model II to be used as an operational model for fitting J-I curves and accurately assess the role of photorespiration on plant photo-protection.