The (effective) elastic thickness of the lithosphere defines the strength of the lithosphere with respect to a load on it. Since the lithosphere is buoyant on a viscous mantle, its behaviour with respect to a load is not fully elastic, but rather viscoelastic. Fennoscandia is a well-known area in the world where the lithosphere has not yet reached its isostatic equilibrium due to the ongoing uplift after the last glacial period at the end of the Pleistocene. To accommodate for this changing property of the lithosphere in time, we present the flexural model of isostasy that accommodates temporal variations of the lithospheric flexure. We then define a theoretical model for computing the elastic thickness of the lithosphere based on combining the flexural and gravimetric models of isostasy. We demonstrate that differences between the elastic and viscoelastic models are not that significant in Fennoscandia. This finding is explained by a relatively young age of the glacial load when compared to the Maxwell relaxation time. The approximation of an elastic shell is then permissible in order to determine the lithospheric structure and its properties. In this way, the elastic thickness can be estimated based on combining gravimetric and flexural models of isostasy. This approach takes into consideration the topographic and ocean-floor (bathymetric) relief as well as the lithospheric structural composition and the post-glacial rebound. In addition, rheological properties of the lithosphere are taken into consideration by means of involving the Young modulus and the Poisson ratio in the model, both parameters determined from seismic velocities. The results reveal that despite changes in the Moho geometry attributed to the glacial isostatic adjustment in Fennoscandia are typically less than 1 km, the corresponding changes in the lithospheric elastic thickness could reach or even exceed ±50 km. The sensitivity analysis confirms that even small changes in input parameters could significantly modify the result (i.e. the elastic thickness estimates). The reason is that the elastic thickness estimation is an inverse problem. Consequently, small changes in input parameters can lead to large changes in the elastic thickness estimates. These findings indicate that a robust estimation of the elastic thickness by our method is possible if comprehensive information about structural and rheological properties of the lithosphere as input parameters are known with a relatively high accuracy. Otherwise, even small uncertainties in these parameters could result in large errors in the elastic thickness estimates.