There is still a grid. If my power source goes out, I want backup to come from other nearby sources - the timeline of power restoration from the current delivery system is on the order of minutes or hours for over 80% of outages, and on the order of a couple weeks for over 99% of the rest of outages. If my power plant breaks, I need restoration numbers that meet that. (additionally, I need plant repair bills to be lower than however much money having the backyard plant would save me. TCO considerations again).
Further, these two assumptions are built into your "better" assessment:
* It is cheaper to have a power plant in my back yard than buying it from the grid.
* The backyard source can be made safe.
Combining these two assumptions is a big deal. If both are true, I will agree that it is a good option (with the caveat listed above). However, there is a HUGE amount of R&D to get there, including a massive set of efficient production runs for parts to build all the systems to make it happen. The economics of this points to it not being likely that everyone has a backyard fusion plant.
It is far more likely to see big fusion plants in greater number scattered around the power grid to provide higher reliability in the cases of line loss etc. Further, with energy now being much, much cheaper to produce, you'll likely start seeing more reliable distribution channels for electrical power. Overhead lines would be reasonable to replace with underground ones, which are less efficient, but are also more reliable as they are less likely to be damaged in weather events. You'll also probably see a reduction in star-topology distribution - more redundancy in distribution paths, at the cost of some efficiency, because the complex equipment will be cheaper to manufacture (you know, because energy to do so will not factor into costs anymore).
