I was just illustrating the path along the lunar gateway (LEO -> TLI -> Gateway -> LLO -> surface). Even if you do a direct (LEO -> TLI -> LLO -> surface) route, you still have to go through LLO as you'll need to be able to pick your landing spot and potentially even delay a landing (which is not possible if LLO is not budgeted in).
Actually now that I think about it, if you want to save delta-v, you can get an extra 25% off the TLI leg of the direct route with a low energy transfer using weak stability boundary trajectories. These savings however come at the cost of a significantly longer flight time and far slimmer margins for failure.
Looking at it from LLO isn't the number you should be using. It's the number from Lunar intercept orbits to landing.