For logic to really go on wheels you need decision procedures that are decidable and for many of the logical systems you would want to use for real-life applications like engineering and bank regulation you want FOL + arithmetic which very much has problems. It's why we get stuck with things like OWL that seem overcomplicated and underpowered at the same time.
Don't get me started on how real commonsense reasoning requires partitioned knowledge bases, modal logic, temporal logic, social logic (Jim thinks that Mary believes ...), all sorts of complications for which a complete and consistent decision procedure is just unthinkable. I think you could put together a logic system for families of problems in certain domains and accomplish a lot but there is no general logic for real life and efforts like Cyc have gotten stuck in the mud or gone down in flames.