If you have vertical lines of pixels on an iPhone painted black and white alternately, then you will have 300 switches from black to white and vice versa in an inch. Thus, he shouldn't claim that you need twice as many pixels for a certain line resolution. You should try doing the math and calculating the angular resolution like I did.

You are right because the article has an error in its computations. In fact, the article uses the value of 50 CPD, which is 1.2 arcminutes. Really good eyes can manage 1 arcminute resolution. Where the author goes wrong is by saying that because a line pair requires two pixels, he can magically make the eye twice as god and say that it has a resolution of 0.6 arcminutes per pixel. This is wrong, as the each pixel is a part of two line pairs. One to its right and one to its left. Thus, the resolution of the eye would actually be 1.2 arcminutes per pixel (or 1 arcminute per pixel using the smaller value).