why are we connecting current node to head then head back to current node?

Question

The code below partitions the LinkedList into a RHS and LHS, and values below the partition (x) falls into the LHS and nodes greater than partition fall onto the RHS.

Now