stored procedure question

I have the following coded for a stored on an oracle server:

CREATE OR REPLACE PROCEDURE NIS.CUR_TEST(p_job_no IN NUMBER,
p_cli_name IN VARCHAR2,
p_error_msg OUT VARCHAR2,
p_ret_val OUT VARCHAR2)
IS
/*********GETS TABLE NAMES/COLUMN NAMES FROM DATABASE******/
CURSOR l_cur IS
SELECT table_name, column_name
FROM cols
WHERE table_name LIKE 'SF_%'
AND column_name = 'JOB_NO';
BEGIN
FOR l_rec IN l_cur LOOP
DECLARE
/*** GETS MAX JOB NUMBER IN EACH TABLE ****/
CURSOR l_cur_table IS
SELECT MAX(job_no) l_job_no
FROM l_rec.table_name;
BEGIN
FOR l_rec_table IN l_cur_table LOOP
/** delete all but the 1000 highest job nos
DELETE
FROM l_rec.table_name
WHERE job_no < l_rec_table.l_job_no - 1000;
COMMIT;
END LOOP;
END;
END LOOP;
END CUR_TEST;
/

The problem is that when I get the table and column names in the first cursor, the next query shown below, the one in the second cursor (l_cur_table) errors:

SELECT MAX(job_no) l_job_no
FROM l_rec.table_name

The error says something about l_rec.table_name not being accessible but really the problem is that I'm trying to use a variable (l_rec.table_name) in the FROM clause. Even if I store l_rec.table_name in a local variable before I get to this query and use that variable in the FROM clause it still errors because it can't interpret a variable value in the FROM clause. Any ideas.

Thanks so much.

Sue