Modal testing is an efficient tool for checking and updating of computational dynamic models of aircraft. Test objects are fixed by special elastic suspension systems during the modal testing time. Stiffness characteristics of such systems are determined on the assumption that the impact of the suspension on the eigentones of elastic vibrations of a freely flying aircraft should not exceed the pre-arranged level. It is therefore considered that the vibration frequency of the structure as a rigid body on the suspension should be several times lower than the natural frequency of the first flexible mode. Besides, different sources recommend various ratios between these frequencies. As to the modal testing of large flexible space structures, the task of creating such a suspension becomes many times more difficult because such structures can have very low natural frequencies of flexible modes. The influence of the suspension system on the accuracy of the aircraft modal testing results manifests itself in two ways. On the one hand, the use of the suspension leads to an increase of all natural frequencies of the test object. On the other hand, the occurrence of modes of the object as a rigid body with non-zero frequencies leads to the aircraft phase resonance frequency shifts which determine the natural frequencies of flexible mode tones. The paper presents a technique for correcting an aircraft computational dynamic model taking into account the suspension system. The correction is based on the fact that the errors in measurement of displacements in modal tests (i.e. eigenmodes as well) are greater than the errors in determining natural frequencies by more than an order of magnitude. The effect of modes of an object as a rigid body on an elastic suspension system on the accuracy of determination of its natural frequencies, generalized masses and generalized damping factors of flexible modes is analyzed.