In this paper, we report on the decoding of phase-based information, from steady-state visual evoked potential (SSVEP) recordings, by means of different classifiers. In addition to the ones reported in the literature, we also consider other types of classifiers such as the multilayer feedforward neural network based on multi-valued neurons (MLMVN), and the classifier based on fuzzy logic, which we especially tuned for phase-based SSVEP decoding. The dependency of the decoding accuracy on the number of targets and on the decoding window size are discussed. When comparing existing phase-based SSVEP decoding methods with the proposed ones, we are able to show that the latter ones perform better, for different parameter settings, but especially when having multiple targets. The necessity of optimizing the target frequencies to the individual subject is also discussed.