In this shot, the wave moves through the shot whilst each word "rides" the waves.

In this shot, the house is constructed in a 3D Space in After Effects, and the camera follows "Looong" off the screen, to the next, which would be the smiley face scene.

In this shot, the camera would initially track along the words "And wouldn't it be nice to live together, in the kind of world where". At that point "Where", the camera pulls out to reveal "We Belong", which turns into a smile and two eyes appear, whilst the head bobs side to side. The camera then zoom spirals in, and the next shot appears.

This is an animated rough of the third style frame I uploaded last week. The eyes should "bob" in time with the accordion.