Post-Translational Modifications (PTMs) play most important roles in the accomplishment of biological processes and molecular functions. It is challenging to identify two PTMs for a tandem mass spectrum. In this paper, we proposed a new algorithm to detect two PTMs with unknown types. First, we constructed a Pair of Peak Set (PPS) which is composed of pairs of peaks that have the highest sum of intensities. Second, we revealed the relationship between PPS and the whole experimental spectrum. Third, a series of logic conditions was proposed to detect PTMs from a MS/MS spectrum. Finally, we used a scoring function to rank the candidate hits. We applied the method to a large MS/MS data set and the experimental results demonstrated that the proposed method achieved better performance of identifying any types of PTMs in a blind mode than current existing methods.