Answer:
Sequential file processing is just working through the input file one record at a time in the sequence that the information has been stored.
'Sequential' is still a kind of file organization.
Large mainframe installations frequently still use tape storage, the data of which is written and read sequentially.
For efficient writing and retrieval of data on disc storage, there are alternatives. Files can be indexed on key fields and retrieval can search the index (quicker than reading each record) which will then direct the program to the correct record within the file.
You might also search the index for all records of a particular type. The indexes are set up while the database is being built. If a new way of seeing the data is required than a new index may need to be built.
You see this every day, for example.. you can store songs on an MP3 player and the MP3 player can play them back to you one after another. There is often data attached to the song when it is stored that tells the genre, date, album name, etc. So you could then ask the player to play only songs from a particular genre or to play everything in chronological sequence. The songs are stored in the MP3 player sequentially but you are playing them back in a different order because you are using a different organization.