On certain rigorous methods of treating problems in celestial mechanics