I've used various methods to size necks, including standard bushing die setups, and expander mandrels. My original thought for using the mandrels was that their use would likely push inconsistencies in neck wall thickness (I don't turn necks) to the outside, and leave them very straight. FWIW - the carbide mandrels available for neck turning purposes are generally held to much tighter tolerances (actual diameter) than the steel ones. Regardless of the price of the Porter die shown previously, the big advantage there is the precision and range of diameters available; something you cannot easily obtain with neck turning mandrels.
My approach with the mandrels has been to use a bushing die first to size the neck, with a bushing that was either .001" or .002" smaller than that which would give me ~.002" neck tension (interference fit). That way, the mandrel would definitely be opening up all the necks in the 2nd step to some degree, presumably allowing for better final diameter consistency than by using a bushing that sized the neck very close to that which the mandrel would open it back up. For .223 Rem and .308 Win, I used .222" and .306" carbide neck turning mandrels, respectively. I typically use a 0.248" bushing with .223 Rem brass (Lapua) and a 0.336" bushing with .308 Win brass (also Lapua). As the first step, I have sized necks with both 0.246" and 0.247" bushings for the .223 brass, and 0.334" and 0.335" bushings with the .308 brass. If you wish to try this method, the size of the bushing used does make a difference.
The point I wanted to bring out with this post, particularly for those that may not have ever tried using a mandrel before, is that sometimes something that seems like a good idea may or may not necessarily offer an advantage, depending on your particular shooting needs. You have to try it to find out. In my hands for the purpose of F-TR shooting, there was really no advantage to using the mandrel. IMO - the necks were not any better or more uniform than I was already achieving using a standard bushing die with half the effort (i.e. one less step). Your results may vary from mine depending on the type of shooting you do.
The other thing I will point out is that you want to take springback into account if you decide to use this approach, especially if you don't have access to an almost unlimited selection of mandrel sizes like the ones from Porter, which I didn't. My thought had been that using a carbide neck turning mandrel of .002" under bullet diameter would give me about the same neck tension that a bushing already known to provide .002" interference fit would give. This was not so, and springback is the reason. When sizing brass necks down from the outside using a bushing, the springback of the brass opens the necks back up just the slightest bit. In contrast, when opening the neck up with a mandrel as the final sizing step, the direction of the springback is opposite; that is, they close up just a tick. The final result is that although you might think the two approaches would give very similar neck tension (at least on paper), in fact they did not. The neck turning mandrel approach gave me between .0005" and .001" greater neck tension than the bushing die already known to give about .002" interference fit. As a result, bullets were noticeably more resistant to seating with the mandrel-prepped necks. Of course, none of that matters if you have access to whatever diameter mandrel you need, as with the Porter setup. However, anyone wishing to attempt this using commercially available neck turning mandrel, just be aware that due to springback, your neck tension will be noticeably greater than about .002", if that's what you have been using previously with a bushing die approach.