In any reliability analysis, the degradation modeling is a key point. Indeed, the accuracy of all reliability indicators and prognosis estimations will directly depends on the quality of the degradation modeling. Commonly used stochastic models such as Markov chains, Gamma process... are generally based on some strong assumptions on the stochastic properties of the considered degradation process that can induce some prejudicial losses of information. In many studies the Dynamic Bayesian Networks formalism (DBN) has been proved relevant to perform reliability studies, since a modeling based on discrete and finite states space is acceptable. In this paper some specific DBN structures will be introduced in order to improve the degradation modeling and perform reliability analysis, integrating the concept of conditional sojourn time distributions that allow considering simultaneously several degradation dynamics. A comparative study with simulated data has been finally carried.