To evaluate the immediate and water-aged microtensile bond strength of a universal adhesive to dentin, using the self-etch approach with and without an additional hydrophobic resin layer. Methods: Flat dentin surfaces were prepared from twelve non-carious human molars and were randomly divided into two groups. The universal adhesive system Scotchbond™ Universal was used in self-etch mode (SBU) with and without the application of an extra hydrophobic resin layer of Adper™ Scotchbond™ MultiPurpose Plus (SBU+HL). After composite buildups, samples were stored in water (37ºC/7 days) and then sectioned into microspecimens (1.00±0.2 mm 2). Half of the microspecimens were immediately subjected to microtensile bond strength testing (0.5 mm/min) while the other half was stored in water, according to the ISO/TS 11405:2015, for 4 years before testing. Data were analyzed with Kruskal Wallis and all-pairwise comparisons with Bonferroni corrections (p<0.05). Results: The following microtensile bond strengths were registered (mean in MPa±SD): immediate SBU, 52.5±13.56; 4-year SBU, 26.0±6.50; immediate SBU+HL, 59.3±12.39; 4-year SBU+HL, 32.5±5.72. No statistically significant differences were detected between adhesive strategies either immediately or in the aged period (p>0.05). A significant decrease in bond strength was verified between the immediate and the 4-year evaluation for both groups (p<0.05). Conclusions: Water storage induced an approximately 50% reduction in dentin bond strength, regardless of the adhesive strategy employed. The incorporation of an extra hydrophobic layer over the universal adhesive did not improve dentin bond strength significantly either immediately or after long-term water storage. (Rev Port Estomatol Med Dent Cir Maxilofac.