A common refrain from the opponents of 345 has been that the majority of the traffic represents drivers going from southeast Dallas to jobs in far north Dallas. This is undoubtedly true. Mostly because the presence of 345 subsidizes and induces that behavior. It appears to be a straight line connection and thus many people end up competing for the same space. This is Braess' Paradox in action, where people acting in their own perceived self-interest (free road, straight line) end up reducing the overall performance and capacity of the road.
However, despite the patronizing, father knows best tone of the 345 opponents pretending to be looking out for the common man, when in actuality they're mere defenders of the status quo, it turns out that 345 is not, nor is ever, the most convenient route from the Pleasant Grove area to Richardson and points north, the direction of the job spillage and inertia out of Dallas.
The above shows a typical google maps suggested routing. Using 345 at this particular time did also have some "normal traffic" at the time where 635 loop had light traffic. However, that doesn't change things significantly when there is even light traffic on 345 as the distance is in fact longer. Further, when is 345 ever actually light traffic? 345 itself is the segment between two choke points next to downtown Dallas where we've induced local and regional traffic to compete for the same space, therefore it's nearly always congested.
Of course, it's also worth noting why that job spillage and inertia is occurring in the first place. Why there aren't jobs closer to downtown and thus closer to South Dallas. Because we subsidize the long trip and the congestion it brings. And we're really listening to the same people that created this inertia and entropy?