In terms of where the water comes from, sourcing at Old River Control makes a lot of sense. From there the water really does pretty much run into the Gulf of Mexico, and there is no downstream irrigation, power generation, or shipping impact from a 5% diversion. But the writer makes the engineering sound trivial, when it's really not. The first thousand miles are all uphill, so you're not talking about a canal, as the author implies, but rather a pipe. And not a small pipe either. If you move the water at a velocity of a meter per second through a single pipe, you're looking at a pipe 40 meters in diameter. And you have to push that against a head of 4000 ft net. Obviously, you'd want to pump at multiple points with intermediate reservoirs to stage the pressure gradient - otherwise you'd have a static pressure of 1800 psi in the pipe at the source. But when you roll up the pipe, the reservoirs, the pumps, and the multiple power plants to run it all, you're talking about a lot more than the garden hose across Texas the author wants to make this sound like.

And much of this is across already arid, water starved land. I'm guessing West Texas and New Mexico will want their own cut of water in return for permitting the project to cut their states in two. You might need to double your total take to satisfy everybody.