The mixing of CBN-printed notes to make bricks of is something we've been seeing for some time. The mixing reported in this thread is pretty normal, and is not the result of insert notes being added.
As one of the researchers who broke the patterns for note printing and numbering, I have a pretty good feel for what is going on. First of all, we know that CBN skip-numbers its sheets by 8,000. That means the interval in serial numbers between chronological notes on a sheet is 8,000. Secondly, we know that sheets are cut in stacks of 100 at a time. That's how patterns in "cup marks" are produced in bundles of 100 notes. What is not 100% clear is how the printers assemble bundles and bricks. Before sheets are cut, defective sheets are taken out and replaced by insert sheets, though the inserts are usually added to the ends of stacks of sheets. A hundred sheets are taken and cut into 45 bundles of 100 singles. These bundles are then put together into bricks of 1,000 notes. That means you'll get up to 10 different positions from the same 100 sheets used to make bricks. Now the part that confuses a lot of people is why there's skipping in the serial numbers within bundles of 8,000 or multiples of 8,000. That could be the result of shuffling before the bundles are assembled, but that would destroy the pattern of cup marks within bundles. I don't have access to bricks, so I can only go by what people who do get bricks on a regular basis describe to me.
What is clear is that you can't have insert notes from the same ream as the rest of the brick. A ream is a grouping of sheets that make up one continuously-numbered sequence of notes. In this case, 45 notes per sheet multiplied by a skip interval of 8,000 equals 360,000 notes in a ream. Inserts have to come from outside that 360,000-note ream to be inserts, otherwise they're just shuffled notes from the same ream.
