A robust and extended characterization of the Point Spread Function (PSF) is crucial to extract the photometric information produced by deep imaging surveys. Here we present the extended PSFs of the Sloan Digital Sky Survey (SDSS), one of the most productive astronomical surveys of all time. By stacking ~1000 images of individual stars with different brightness, we obtain the bidimensional SDSS PSFs extending over 8 arcmin in radius for all the SDSS filters (u, g, r, i, z). This new characterization of the SDSS PSFs is near a factor of 10 larger in extension than previous PSFs characterizations of the same survey. We found asymmetries in the shape of the PSFs caused by the drift scanning observing mode. The flux of the PSFs is larger along the drift scanning direction. Following a reproducible science philosophy, we make all the PSF models and the used tools publicly available. Finally, we illustrate with an example how the PSF models can be used to remove the scattered light field produced by the brightest stars in the Coma Cluster central region. This particular example shows the huge importance of PSFs in the study of the low surface brightness Universe, especially with the upcoming of ultra-deep surveys such as the Large Synoptic Survey Telescope (LSST).