It is difficult to accurately predict the flexural strength of prestressed members with unbonded tendons, unlike that of prestressed members with bonded tendons, due to the unbonded behavior between concrete and tendon. While there have been many studies on this subject, the flexural strength of prestressed members with unbonded tendons is still not well understood, and different standards in various countries often result in different estimation results for identical members. Therefore, this paper aimed to observe existing approaches and to propose an improved model for the ultimate strength of prestressed members with unbonded tendons. Additionally, a large number of tests results on flexural strength of prestressed members with unbonded tendons were collected from previous studies, which entered into a database to verify the accuracy of the proposed model. The proposed model, compared to existing approaches, well estimated the flexural strength of prestressed members with unbonded tendons, adequately reflecting the effects of influencing factors such as the reinforced steel ratio, the loading patterns, and the concrete strength. The proposed model also provided a reasonably good estimation of the ultimate strength of over-reinforced members and high-strength concrete members.