In this paper we provide an overview of fuzzy modeling methods applied to time series processing. The basic methods are Fuzzy Transform (F-transform) and selected methods of Fuzzy Natural Logic (FNL). We address classical tasks such as estimation of trend and its prediction, and also methods for mining information from time series. We provide information that can hardly be obtained using statistics. Namely, we automatically form an explanation of the forecast in natural language, provide comments to the slope of time series in an imprecisely specified area, detect possible structural breaks, "bull and bear" phases of financial time series, measure of similarity between time series and provide automatic summarization of knowledge about time series expressed in natural language.