Essentially this is giving you the name of the row with the largest area for each continent.
Line 1 is your final result set from world x.
Line 2 is filtering your final result set to the row with the largest area.
Line3/4 is your subquery which is being used for the comparison. This subquery will be executed for each row in the dataset.
Because the where clause of the subquery is ensuring that the continents are equal each row is only compared to rows from the same continent.
So for each row in x it is compared to each row in y where the continents match. If the row is >= every row in y it ends up in your final result set.
(this is to help understanding and not necessarily exactly how the work is performed by the sql engine)
For each row in the primary query the subquery is run and returns all rows that match the continent of the current row in the primary query. Comparison is made for the primary row to each subquery row and the final result set is populated with only the row from the primary where area is >= to all of the rows in the subquery. It then moves on to the second row of the primary query until all comparisons have been made and then returns the final result set to you.